GPU 서버 중앙 AI 라우팅 서비스 초기 구현: - OpenAI 호환 API (/v1/chat/completions, /v1/models, /v1/embeddings) - 모델 레지스트리 + 백엔드 헬스체크 (30초 루프) - Ollama SSE 프록시 (NDJSON → OpenAI SSE 변환) - JWT 인증 이중 경로 (httpOnly 쿠키 + Bearer 토큰) - owner/guest 역할 분리, 로그인 rate limiting - 백엔드별 rate limiting (NanoClaude 대비) - SQLite 스키마 사전 정의 (aiosqlite + WAL) - Docker Compose + Caddy 리버스 프록시 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
80 lines
1.9 KiB
Python
80 lines
1.9 KiB
Python
from fastapi import APIRouter, Request, Response
|
|
from pydantic import BaseModel
|
|
|
|
from config import settings
|
|
from middleware.auth import (
|
|
check_login_rate_limit,
|
|
create_token,
|
|
record_login_attempt,
|
|
)
|
|
|
|
router = APIRouter(prefix="/auth", tags=["auth"])
|
|
|
|
|
|
class LoginRequest(BaseModel):
|
|
password: str
|
|
|
|
|
|
class LoginResponse(BaseModel):
|
|
role: str
|
|
token: str
|
|
|
|
|
|
@router.post("/login")
|
|
async def login(body: LoginRequest, request: Request, response: Response):
|
|
ip = request.client.host if request.client else "unknown"
|
|
|
|
if not check_login_rate_limit(ip):
|
|
return _error_response(429, "Too many login attempts. Try again in 1 minute.")
|
|
|
|
record_login_attempt(ip)
|
|
|
|
if body.password == settings.owner_password:
|
|
role = "owner"
|
|
elif body.password == settings.guest_password:
|
|
role = "guest"
|
|
else:
|
|
return _error_response(401, "Invalid password")
|
|
|
|
token = create_token(role)
|
|
|
|
# Set httpOnly cookie for web UI
|
|
response.set_cookie(
|
|
key="token",
|
|
value=token,
|
|
httponly=True,
|
|
samesite="lax",
|
|
max_age=settings.jwt_expire_hours * 3600,
|
|
)
|
|
|
|
return LoginResponse(role=role, token=token)
|
|
|
|
|
|
@router.get("/me")
|
|
async def me(request: Request):
|
|
role = getattr(request.state, "role", "anonymous")
|
|
if role == "anonymous":
|
|
return _error_response(401, "Not authenticated")
|
|
return {"role": role}
|
|
|
|
|
|
@router.post("/logout")
|
|
async def logout(response: Response):
|
|
response.delete_cookie("token")
|
|
return {"ok": True}
|
|
|
|
|
|
def _error_response(status_code: int, message: str):
|
|
from fastapi.responses import JSONResponse
|
|
|
|
return JSONResponse(
|
|
status_code=status_code,
|
|
content={
|
|
"error": {
|
|
"message": message,
|
|
"type": "auth_error",
|
|
"code": f"auth_{status_code}",
|
|
}
|
|
},
|
|
)
|